A hallmark of cancer is that it is a disease of gene misexpression. This application addresses how gene expression in cancer is disrupted by aberrant maturation of mRNA 3' ends where the poly (A) tail is added. Most human genes have multiple poly (A) sites and, as a result, making choices between alternative poly (A) sites is an almost ubiquitous event in gene expression. In recent groundbreaking work by other labs, it was reported that alternative poly (A) site choice is commonly disrupted in cancer cells. This is important because poly(A) site choice can have a large impact on protein expression by dictating what sequences are included in an mRNA's 3' untranslated region (UTR). 3' UTR sequences are the principle targets for microRNAs and RNA binding proteins that regulate mRNA stability and translation efficiency. For example mRNAs will evade these controls if a poly (A) site is chosen that cuts off these target sequences to produce an abbreviated 3'UTR. We still lack comprehensive information about which genes are affected by abnormal poly(A) site selection in cancer cells and whether this process is regulated during tumor evolution. Furthermore little is known about the basis for how one poly(A) site is selected over another for processing. What might be going wrong in cancer cells to divert mRNA 3' end processing from the correct poly (A) sites is very much a mystery. We propose to tackle two broad fundamental questions: 1) What genes are affected by aberrant poly(A) site choice in cancer cells during tumor progression, adaptation to hypoxia, and the response to therapeutics? and 2) What is the molecular basis for the corruption of mRNA 3' end formation in cancer cells?. We will use a genome-wide RNA-seq and ChIP-seq strategies to answer these questions using human breast and lung cancer cells.